J. Agric. Food Chem. 2008, 56, 1829–1837
1829
Determination of Trifloxystrobin and Its Metabolites in Hawaii Soils by ASE-LC-MS/MS JINGYU CHEN,*,†,‡ BINH LOO,‡
AND
CHITTARANJAN RAY†,§
Water Resources Research Center and Department of Civil and Environmental Engineering, University of Hawaii at Manoa, 2540 Dole Street, Honolulu, Hawaii 96822, and Chemical Analysis Laboratory, Hawaii Department of Agriculture, 2725 Waimano Home Road, Pearl City, Hawaii 96782
Analytical methods for the determination of trifloxystrobin and four of its metabolites were developed in a leaching study conducted in Hawaii. To duplicate plots at each of five locations representing various agricultural areas in Hawaii, trifloxystrobin was applied at label rates and allowed to leach under normal rain and irrigation conditions. Soil samples were collected at weekly to monthly intervals and the residual concentrations of trifloxystrobin and metabolites measured. A quantitative analytical method for their determination in various soil samples was developed using accelerated solvent extraction (ASE), coupled with liquid chromatography-tandem mass spectrometry. Extraction solvent with various ratios of methanol to water, addition of EDTANa2 to the extract solvent, and ASE cell temperature were adjusted to improve recovery. Deuterated (E,E)-trifloxystrobin was chosen as the internal standard of the analytical method. The limit of quantitation was 2.5 ppb in the soil for trifloxystrobin and its metabolites. Laboratory aerobic degradation studies with the soils from the five sites were also conducted to measure the same compounds. KEYWORDS: Trifloxystrobin; trifloxystrobin acid; LC-MS/MS; ASE
INTRODUCTION
Trifloxystrobin is a mesostemic fungicide that interferes with the respiration of plant-pathogenic fungi. It is mostly used as a broad-spectrum foliar fungicide against many fungal pathogens within the Ascomycete, Deuteromycete, Basidiomycete, and Oomycete classes. In particular, it is used to treat powdery mildew, downy mildew, and anthracnose, which mostly attack fruit and vegetable crops (1). The compound has been recently introduced in Hawaii for fungus control in greenhouse and nursery crops and for field crops. Trifloxystrobin (an ester) and its hydrolyzed acid metabolite (Figure 1) have four geometrical isomers based on the connected group arrangement of the two NdC double bonds. Their acronyms, CAS Registry Numbers, and chemical names are presented in Table 1. (E,E)-Trifloxystrobin is the most biologically active fungicide among the four isomers. Trifloxystrobin degrades rapidly in water and soil by mechanisms including metabolism, photolysis, and hydrolysis. Banerjee et al. (2, 3) showed that soil hydrolysis was the dominant pathway of degradation of trifloxystrobin and that the major metabolite was the E,E acid. Isomerization was of minor importance due to the absence of light. The same studies demonstrated that * Corresponding author [telephone (808) 453-5923; fax (808) 4535924; e-mail
[email protected]]. † Water Resources Research Center, University of Hawaii at Manoa. ‡ Hawaii Department of Agriculture. § Department of Civil and Environmental Engineering, University of Hawaii at Manoa.
Figure 1. Chemical structures of (E,E)-trifloxystrobin and (E,E)-trifloxystrobin acid.
adsorption of both the parent compound and its metabolite acids follow linear and Freundlich isotherms; the extent of adsorption depends upon the binding sites available in organic matter and clay (2, 3). Reported analytical methods for trifloxystrobin are primarily those for residue analysis from viniculture or other fruit and vegetable crops (4–8). Most literature reports the use of a variety of organic solvents for extraction; however, the literature reports limited use of supercritical fluid extraction (9). Solid-phase extraction (SPE) with various types of cartridges (6, 10, 11), matrix solid-phase dispersion (5, 12), gel permeation chromatography (13), and liquid–liquid extraction (4, 8) have all been used to purify crude extracts. Following purification, either liquid chromatography (LC) coupled with mass spectrometry (MS) (11) or with diode array detection (4, 8) or gas chromatography
10.1021/jf071527z CCC: $40.75 2008 American Chemical Society Published on Web 02/20/2008
1830
J. Agric. Food Chem., Vol. 56, No. 6, 2008
Chen et al.
Table 1. Chemical Names, Acronyms, CAS Registry Numbers, and Codes of Trifloxystrobin and Its Metabolites chemical name
acronym
CAS Registry No.
code
(E,E)-trifloxystrobin (Z,Z)-trifloxystrobin (Z,E)-trifloxystrobin (E,Z)-trifloxystrobin (E,E)-trifloxystrobin acid (Z,Z)-trifloxystrobin acid (Z,E)-trifloxystrobin acid (E,Z)-trifloxystrobin acid
EETF ZZTF ZETF EZTF EETFA ZZTFA ZETFA EZTFA
141517-21-7
CGA279202 CGA357262 CGA357261 CGA331409 CGA321113
141493-03-0 252913-85-2
CGA373466
using a nitrogen-phosphorus detector (6, 8, 9), electron capture detector (6, 8), and MS detector (5, 8, 14) has been used for analysis. The former owner and developer of the product, Novartis Crop Protection, established two methods to analyze trifloxystrobin and its metabolites from soil (15, 16). The most frequently used method involves extraction with 90% methanol (by shaking), followed by filtration and evaporation. The sample residue is further cleaned by passing it through a series of SPE cartridges (mixed-phase cation exchange disk cartridge, aminopropyl extraction disk cartridge, and a silica gel cartridge) before it is analyzed by LC-UV (15). The analytical method was subsequently improved using reversed phase LC coupled with a single-quadrupole mass detector (or LC-MS). The selectivity of the mass spectrometer eliminates the need to use SPE for sample cleanup. After the solvent is changed from 90% methanol to 50% acetonitrile, the solution is directly injected into the LC-MS. The limit of quantitation (LOQ, defined as the lowest fortification that still gave adequate recovery) is 10 parts per billion (ppb) in soil (16). Following the modified method of the developer (15, 16), poor recovery (20–30%) was obtained for the trifloxystrobin acids, which might be due to the different physicochemical properties of the tropical soils of Hawaii. Finally, we successfully implemented accelerated solvent extraction (ASE) for extraction and high-performance liquid chromatography (HPLC) with tandem mass spectrometry (LC-MS/MS) for the analyses of trifloxystrobin and its metabolites. To our knowledge, this is the first time ASE has been employed to extract trifloxystrobin and its metabolites from soil. MATERIALS AND METHODS Reagents and Materials. Seven analytical standards obtained from Bayer CropScience (Kansas City, MO) included ZZTFA (95% purity), EETFA (99.1% purity), ZETFA (95% purity), ZETF (97.6% purity), EETF (98.9% purity), ZZTF (99.4% purity), and d3-EETF (1.34%, liquid solution). The parent trifloxystrobin (EETF) and four metabolites (ZZTFA, EETFA, ZETF, and ZZTF) were used in all of the analyses. Later, ZETFA was added to compare its behavior during LC-MS analysis. All analytes were dissolved in methanol as stock solution (1 mg mL-1) and stored in a refrigerator at 4 °C, from which the lowlevel spike solution and the solutions for the calibration curve were prepared by appropriate dilution with 1:1 methanol/water. The spike solution for the high-level recovery test (300 µg mL-1) was prepared by directly dissolving 3 mg of each five target compounds in 10 mL of methanol. The internal standard solution was prepared by diluting the stock solution of d3-EETF to 0.5 µg mL-1 with 1:1 methanol/ water. To prepare the standard solutions for calibration curve (0.5–10 ng mL-1), various volumes of the 1 µg/mL mixture (of five target compounds) standard solution were added into a 50 mL volumetric flask, followed by the addition of 25 mL of extraction solvent (50% methanol and 50% water, contained 0.5 g L-1 EDTANa2), the volume of which was similar to the ASE extract volume. Subsequently, 1 mL of internal standard (0.5 µg mL-1) was added to each flask. The final solution volumes were adjusted to 50 mL with 1:1 methanol/water.
Figure 2. Experimental sites for measuring pesticide leaching evaluation in the soils of Hawaii. HPLC-grade methanol was purchased from Burdick & Jackson (Muskegon, MI). Certified ACS-grade ethylenediaminetetraacetic acid disodium salt-2-hydrate (EDTANa2) and ammonium formate were purchased from Fisher Scientific (Houston, TX). Deionized water was processed through a Millipore ultra pure system (Milli-Q, Millipore, Milford, MA). Diatomaceous earth as well as purified, washed, and ignited sand was purchased from J. T. Baker (Phillipsburg, NJ). Sites and Soils. Poamoho, Waimanalo, and Kunia on the island of Oahu; Kula on the island of Maui; and Mana on the island of Kauai were selected for study because they offer differing soil, climatic, and topographic characteristics representative of agriculturally important soils in Hawaii. The locations of these five sites are presented in Figure 2. The soils chosen for this study differ in moisture and temperature regimens, mineralogy, and physical and chemical properties. The test plots are located on well-drained sites, flat or gentle slopes, or level terraces where erosion would be minimal. The physicochemical properties of the soils from these five locations are shown in Table 2. Accelerated Solvent Extraction. ASE 200 (Dionex, Sunnyvale, CA) was used for soil sample extraction. Preweighed samples (10 g) were stored in a freezer. Prior to analysis, the samples were brought to room temperature. Soil samples with various moisture contents were mixed with diatomaceous earth and then poured into 22-cm extraction cells. The extraction cell of ASE was filled to the top with sand. The extraction solution was a 50:50 mixture of methanol/water; the water contained 0.5 g L-1 of EDTANa2. The oven temperature was set at 40 °C and cell pressure at 10.24 MPa. Solvent was introduced and heated to the set temperature in 5 min, followed by another 5 min of static time (solvent soaking time) of extraction. An amount of solvent equivalent to 60% of the cell volume was flushed through the cell. After each extraction, the system was purged with nitrogen for 60 s at 1.035 MPa. The extraction cycle per cell is 1. The extracts were transferred from the collection vials to 50-mL volumetric flasks. The collection vials were washed two times with 1 mL of methanol, and then the wash solutions were added to the volumetric flasks. One milliliter of internal standard (0.5 µg mL-1) was added to each flask. The final solution volumes were adjusted to 50 mL with 1:1 methanol/ water. Twenty microliters of each of these solutions was injected into LC-MS/MS for analysis. Liquid Chromatography and Tandem Mass Spectrometry. A Thermo Finnigan Surveyor HPLC system (Waltham, MA) equipped with a Phenomenex (Torrance, CA) Gemini MS C18 column (2.0 mm × 150 mm, 5 µm) was employed in the analysis of trifloxystrobin and its metabolites. The mobile phase was a 30:70 (v/v) mixture of 10 mM aqueous ammonium formate and methanol. The flow rate was 0.2 mL min-1, and the total run time was 15 min. The injection volume was 20 µL. The calibration range was 0.5–10 ng mL-1 for all compounds. A Thermo Finnigan LCQ ion trap mass spectrometer equipped with an electrospray ionization (ESI) source was used as the HPLC detector. A metal needle system was installed in the ESI source. The capillary temperature was 250 °C, and the respective flow rates of the sheath gas and auxiliary gas were 80 and 15 (arbitrary units). The spray voltage was 4.5 kV. The two-stage full scan mode and the positive ion (PI) mode were used. The [M + H]+ ions of the five compounds were selected for
Trifloxystrobin and Its Metabolites in Hawaii Soils
J. Agric. Food Chem., Vol. 56, No. 6, 2008
1831
Table 2. Physicochemical Properties of Soils From Five Locations in Hawaii Poamoho, Oahu (P)
Waimanalo, Oahu (W)
Kunia, Oahu (K)
Kula, Maui (M)
Mana, Kauai (L)
silty clay Oxisol Typic Eutrotorrox 85 600 historically, sugar cane; now, crops (vegetables, seed corn, etc.) 1–3 1.2–1.3 30–45
loam Andisols Typic Eurtandepts 911 750 pastureland, vegetable and protea floriculture
stony silty clay loam Vertisol Typic Haplusterts 3 500 corn, sunflower, Cucurbitaceae family fruits such as melon or pumpkin 1–3 1.2–1.5 35–47
soil type class elevation (m) annual rainfall (mm) production
silty clay Oxisol Rhodic Eutrustox 213 1000 pineapple
slope (%) bulk density (g/cm3) field capacity water contenta (% volume) organic carbonb (%) cation exchange capacity (CEC)c pHb (standard deviation) mineralogy
1–3 1.2–1.3 40–50
gravelly clay Vertisol Vertic Haplustolls 30 1500 floriculture and crops (banana, papaya, macadamia nuts, corn) 1–3 1.0–1.3 45–55
0.7 (0.5) low
0.9 (0.9) moderate
0.8 (0.4) low
5.5 (0.5) moderate-high
1.0 (0.6)
6.2 (0.1) kaolin/sesquioxide
6.1 (0.2) montmorrilin with kaolin
6.7 (0.1) kaolin/sesquioxide
7.0 (0.2) montmorrilin
8.4 0.1 kaolin/amorphous
1–3 0.7–0.9 35–45
Extrapolated from retention curves at -330 cm suction. b Based on seven samples to a depth of 2.10 m for P, W, and K; four samples to a depth of 0.9 m for M; and six samples to a depth of 1.80 m for L. c The CEC values are based on values from earlier studies. a
Figure 3. LC-ESI/MS chromatograms in positive ion mode (CID ) 28%) of a combined solution (10 ng L-1) of (Z,Z)-trifloxystrobin acid (ZZTFA) (scan
range, m/z 110–396; transition, m/z 395–148), (E,E)-trifloxystrobin acid (EETFA) (scan range, m/z 110–396; transition, m/z 395–186), (Z,E)-trifloxystrobin (ZETF) (scan range, m/z 110–410; transition, m/z 409–206), (E,E)-trifloxystrobin (EETF) (scan range, m/z 110–410; transition, m/z 409–186), d3-(E,E)trifloxystrobin (EETF) (scan range, m/z 110–415; transition, m/z 412–186), and (Z,Z)-trifloxystrobin (ZZTF) (scan range, m/z 110–410; transition, m/z 409–206). monitoring. For all compounds, the capillary voltage was 20 V and the tube lens offset was -15 V. The normalized collision energy, activation Q, and activation time were tuned on the basis of the selected daughter ion current. For all analytes, the collision-induced dissociation (CID) was 28%, activation Q was 0.25, and activation time was 30 ms. Unlike the selected reaction monitoring (SRM), the two-stage fullscan mode is a two-step MS procedure monitoring the full product ions, which yield more information than that of SRM mode. However, in the software processing method, only the most prominent product ions of the studied compounds were chosen for quantitation in this study, which made it act as a normal SRM method in the quantitative analysis. Figure 3 shows the chromatographs of the mixture of five compounds at concentrations of 10 µg L-1. The instrument’s limit of detection (LOD, defined as S/N ) 3) was 0.1 µg L-1 for all five compounds. All analyses were done in triplicate. RESULTS AND DISCUSSION
Optimization of ASE Parameters. The composition and temperature of the ASE extract were tuned to maximize recovery
from soil. A low-level spiking solution (1 µg mL-1 of the five compounds’ mixture) was used for this series of analyses. Extraction solvent with different ratios of methanol/water was tested to obtain the best recovery for trifloxystrobin (i.e., EETF) and two of its acid metabolites (i.e., EETFA and ZZTFA). The results in Table 3 show that the higher the methanol/water ratio, the higher the recovery for EETF and the lower the recoveries for ZZTFA and EETFA. This result is consistent with the fact that the acid is more hydrophilic than the ester and thus more likely to be extracted by solutions with a higher water content. A mixture of 1:1 methanol/water was chosen as the extract solvent for all subsequent analyses to increase the recovery of the acids while not excessively reducing the recovery of the ester. Using the 1:1 methanol/water mixture as the solvent, recovery of trifloxystrobin acid was improved for the W soil. However, it was still low for the K soil (Table 4). Because the K soil is rich in Fe3+ and Al3+, it is assumed that a complex was formed
1832
J. Agric. Food Chem., Vol. 56, No. 6, 2008
Chen et al.
Table 3. Mean Recoveries and Relative Standard Deviations (RSD) of Trifloxystrobin (EETF) and Its Two Acid Metabolites, (Z,Z)-Trifloxystrobin Acid (ZZTFA) and (E,E)-Trifloxystrobin Acid (EETFA), at Spiked Level of 0.05 mg kg-1, Obtained by ASE at 100 °C with Different Ratios of Methanol from Waimanalo (W) Soila
Table 6. Base Peak Pattern of Trifloxystrobin Acid, Trifloxystrobin, and Their Isomers, (Z,Z)-Trifloxystrobin Acid (ZZTFA), (Z,E)-Trifloxystrobin Acid (ZZTFA), (E,E)-Trifloxystrobin Acid (EETFA), (Z,E)-Trifloxystrobin (ZETF), and (Z,Z)-Trifloxystrobin (ZZTF), after LC-ESI/MS m/z 148
% methanol
a
EETF
ZZTFA
EETFA
50
recovery (%) RSD
69.7 3.4
79.1 3.9
87.6 1.5
75
recovery (%) RSD
78.9 7.5
53.8 4.9
60.7 3.9
90
recovery (%) RSD
90.6 4.2
39.5 4.8
47.1 7.0
Triplicate tests were done.
Table 4. Mean Recoveries and Relative Standard Deviations (RSD) of Trifloxystrobin (EETF) and Its Metabolites, ZZTFA, EETFA, ZETF, and ZZTF, at a Spiking Level of 0.05 mg kg-1, Obtained by ASE at 40 °C without and with Addition of EDTANa2 in the Extraction Solvent From Kunia (K) Soila extraction solvent
EETF
ZZTFA
EETFA
ZETF
ZZTF
1:1 MeOH/H2O
recovery (%) RSD
78.2 5.6
59.2 7.0
57.7 10.5
78.4 2.6
74.8 4.0
1:1 MeOH/H2O (EDTANa2 0.05%)
recovery (%) RSD
74.8 3.4
78.9 6.6
91.8 2.9
80.5 5.6
75.2 5.3
a
Triplicate tests were done.
Table 5. Mean Recoveries and Relative Standard Deviations (RSD) of Trifloxystrobin and Its Metabolites, (Z,Z)-Trifloxystrobin Acid (ZZTFA), (E,E)-Trifloxystrobin Acid (EETFA), (Z,E)-Trifloxystrobin (ZETF), and (Z,Z)-Trifloxystrobin (ZZTF), at a Spiking Level of 0.05 mg kg-1, Obtained by ASE with Different Oven Temperatures from Kula (M) and Mana (L) Soils Using 50% Methanol and 50% Water (with 0.5 g L-1 of EDTANa2) as Extraction Solventa
soil
ASE oven temp (°C)
EETF
ZZTFA
EETFA
ZETF
ZZTF
L
40
recovery (%) RSD
72.3 4.0
62.5 2.5
69.5 2.4
73.5 2.3
68.6 1.9
70
recovery (%) RSD
90.8 2.0
69.7 2.2
77.9 7.9
85.7 1.0
87.6 4.2
100
recovery (%) RSD
95.7 3.2
70.4 10.6
83.9 2.1
89.2 1.0
93.9 0.4
40
recovery (%) RSD
90.8 4.0
77.3 6.0
68.5 3.1
90.0 1.4
82.5 3.7
70
recovery (%) RSD
87.1 5.4
63.1 14.0
74.5 6.2
89.0 1.1
83.5 2.0
100
recovery (%) RSD
97.8 1.5
67.0 5.3
77.2 5.7
96.0 2.6
97.4 3.5
M
a
Triplicate tests were done.
between the metal ion and the carboxylic and amine groups of the acid metabolites, which led to the strong adsorption of these compounds to the soil and low recovery during extraction. This hypothesis was further supported by the fact that recovery was increased by adding EDTANa2 to the extraction solvent, a summary of which is presented in Table 4. Therefore, 1:1 methanol/water (EDTANa2 0.05%) was finally set as the extraction solvent for the analysis of all soil types.
trifloxystrobin acid trifloxystrobin
ZZTFA, ZETFA
m/z 186 EETFA EETF
m/z 206 ZETF, ZZTF
The effect of ASE oven temperature on recovery is shown in Table 5. No significant difference in recovery under different temperatures was observed in the M soils. For the L soil, a higher temperature resulted in higher recovery; therefore, the ASE solvent should be heated to 100 °C for better recovery from this soil. The L soil is quite different from the other soils. It is quite heterogeneous and mixed with lacustrine clay and large pieces of coral. Possibly, it offers more binding sites. However, during the first several of batches of the sample analysis, instrument malfunction prevented heating to 100 °C and the tests were done under 40 °C. To keep the same extract condition, the whole extraction process for all soil samples in this project was conducted at 40 °C. Chromatographic Separation. Under the HPLC conditions specified above, ZZTFA and ZETFA were not separable. Efforts to separate them failed, even after the proportion of H2O in the mobile phase had been increased to elongate the retention time. The major fragments in the mass spectrum were also similar. Thus, these two isomers could not be differentiated by the technique employed in this study. Only ZZTFA was used to make a standard curve to quantitatively analyze the field samples. Thus, the calculated amount of ZZTFA should include ZETFA (which coelutes with ZZTFA). Because the MS responses of ZZTFA and ZETFA are similar, the results of ZZTFA are actually the sum of the amount of both ZZTFA and ZETFA, which were calculated on the basis of ZZTFA’s calibration curve. Mechanism of Fragmentation. The mass spectroscopic data of trifloxystrobin and its isomers have been discussed by Banerjee et al. (17, 18). Figure 4 shows the mass spectra of trifloxystrobin, trifloxystrobin acid, and their isomers using the ESI source. Table 6 summarizes the isomers’ base peak pattern. Although the previous work stated that the double bond NdC connected to the meta-substituted benzene ring controls the type of base peak (17, 18), it is only consistent with the EI mass spectra. APCI/MS/MS spectra of acid metabolites showed the other pattern: the conformation on the other double bond -NdC connected to the ortho-substituted benzene ring played a more dominant role on the type of base peak, by which EZTFA and ZZTFA have m/z 148 as the base peak, whereas EETFA and ZETFA have 186 as the base peak (18). However, with our ESI source, the NdC connected to the meta-substituted aromatic system still controls the base peak type of both esters and acids. Both EETF and EETFA’s base peak are m/z 186. ZETF and ZZTF have 206 as base peak; whereas their acids, ZETFA and ZZTFA, had less stability in m/z 192 (corresponding to their ester’s m/z 206) and continued to lose CO2 and m/z 148 (192 - CO2) became their base peak. Ion m/z 148 could be further broken down to its daughter ion m/z 116 under MS/MS/MS fragmentation. From the above discussion, the fragmentation mechanism is illustrated in Figure 5. Matrix Effect. It is a common perception that matrixinduced coelutes are likely to be responsible for the analytes’ signal suppression (19). Blank (nonspiked) extract solutions of the five soils were used as a matrix to prepare a 10 µg
Trifloxystrobin and Its Metabolites in Hawaii Soils
J. Agric. Food Chem., Vol. 56, No. 6, 2008
1833
Figure 4. Mass spectrum under two-stage full-scan mode and three-stage full-scan mode by LC-MS/MS of (a) (Z,Z)-trifloxystrobin acid, (b) (Z,E)trifloxystrobin acid, (c) (E,E)-trifloxystrobin acid, (d) (Z,E)-trifloxystrobin, (e) (E,E)-trifloxystrobin, (f) d3-(E,E)-trifloxystrobin, and (g) (Z,Z)-trifloxystrobin.
L-1 standard solution. The peak areas of trifloxystrobin and its metabolites in these solutions were compared to those in
a standard solution prepared with deionized water. The results indicate that moderate ion suppression occurred, leading to
1834
J. Agric. Food Chem., Vol. 56, No. 6, 2008
Chen et al.
Figure 5. Fragmentation mechanism of trifloxystrobin and its acid. Table 7. Mean Peak Area Ratios of Standard Solution Prepared by Matrix Compared to That Prepared with Deionized Water after LC-ESI/MSa soil
a
ZZTFA
EETFA
ZETF
EETF
ZZTF
d3-EETF
K
peak area ratio (K/H2O %) RSD (%)
88.6 1.7
71.7 6.8
93.3 5.8
73.5 2.2
93.1 1.5
88.6 3.4
L
peak area ratio (L/H2O %) RSD (%)
89.0 5.0
85.5 5.7
92.9 8.3
74.3 3.8
90.1 9.4
80.7 4.4
M
peak area ratio (M/H2O %) RSD (%)
83.2 5.5
76.9 5.7
92.4 4.3
72.4 1.5
96.0 5.4
88.1 1.0
P
peak area ratio (P/H2O %) RSD (%)
80.7 1.2
76.5 8.7
93.5 8.1
72.2 2.0
95.6 0.5
82.9 2.3
W
peak area ratio (W/H2O %) RSD (%)
89.4 4.5
88.4 2.9
90.6 4.2
83.9 3.8
90.6 2.4
95.9 4.9
Triplicate tests were done.
Trifloxystrobin and Its Metabolites in Hawaii Soils
J. Agric. Food Chem., Vol. 56, No. 6, 2008
Table 8. Mean Recoveries and Relative Standard Deviations (RSD) of Trifloxystrobin (EETF) and Its Metabolites, ZZTFA, EETFA, ZETF, and ZZTF, in the Recovery Experimentsa soil
EETF
ZZTFA
EETFA
ZETF
ZZTF
73.5 2.3
68.6 1.9
–1
Low Level (0.05 mg kg ) 72.3 62.5 69.5 4.0 2.5 2.4
L
recovery (%) RSD
W
recovery (%) RSD
86.3 9.2
75.2 8.3
80.5 5.7
103.6 3.1
94.7 2.9
M
recovery (%) RSD
90.8 4.0
77.3 6.0
68.5 3.1
90.0 1.4
82.5 3.7
K
recovery (%) RSD recovery (%) RSD
74.8 3.4 71.3 6.0
78.9 6.6 75.0 7.2
91.8 2.9 80.2 7.1
80.5 5.6 91.4 5.1
75.2 5.3 96.6 3.5
L
recovery (%) RSD
High Level (3 mg kg–1) 54.2 60.0 63.5 2.0 7.0 8.0
57.6 1.0
53.5 4.0
W
recovery (%) RSD
74.6 4.0
70.0 6.0
71.6 4.0
77.6 3.0
73.3 3.0
M
recovery (%) RSD
79.1 4.0
75.5 3.0
71.4 4.0
83.1 2.0
80.2 4.0
K
recovery (%) RSD
102.8 1.0
98.7 5.0
107.6 6.0
99.8 4.0
98.6 2.0
P
recovery (%) RSD
91.0 6.0
83.7 2.0
102.5 5.0
94.4 3.0
88.1 2.0
P
a
Triplicate tests were done.
Table 9. Calibration Curves’ Linearity Information of (Z,Z)-Trifloxystrobin Acid, (E,E)-Trifloxystrobin Acid, (Z,E)-Trifloxystrobin, (E,E)-Trifloxystrobin, and (Z,Z)-Trifloxystrobin Generated by LC-ESI/MS/MS ZZTFA
EETFA
ZETF
EETF
ZZTF
slope 0.00897113 0.0102025 0.152802 0.0785677 0.0985367 0.9967 0.9987 0.9986 0.9925 0.9983 correlation coefficient (r2)
Table 10. Concentration (Nanograms per Gram) Profile of EETF and Its Metabolites from the Two Test Plots on Week 16 at the Poamoho (P) Site sampling depth (m)
EETF
ZZTFA + ZETFA
EETFA
ZETF
ZZTF
0–0.15 (1)a 0–0.15 (2) 0.15–0.3 (1) 0.15–0.3 (2) 0.3–0.6 (1) 0.3–0.6 (2) 0.6–0.9 (1) 0.6–0.9 (2) 0.9–1.2 (1) 0.9–1.2 (2) 1.2–1.5 (1) 1.2–1.5 (2) 1.5–1.8 (1) 1.5–1.8 (2) 1.8–2.1 (1) 1.8–2.1 (2)
6.7 4.3 4.3 6.1 4.1 3.2 2.7 3.1 4.8 4.7 3.3 4.0 5.1 2.9 1.3 4.5
NDb ND 1.0 ND ND ND ND ND ND 4.6 ND 1.9 ND ND ND ND
19.5 65.4 5.9 2.0 0.6 ND ND ND 15.3 ND 7.9 1.5 1.1 ND ND ND
ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND
ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND
a
Numbers 1 and 2 in parentheses imply duplicate samples. b Not detected.
the observation that the peak areas of the compounds prepared using the matrix solution decreased to 70–96% of those prepared using deionized water (Table 7). In the quantitative analysis, the peak area ratio of the target compounds to the
1835
Table 11. Concentration (Nanograms per Gram) of EETF and Its Metabolites from Laboratory Degradation at Week 16 for Three Poamoho Soils Collected from Depth Intervals of 0–0.15, 0.15–0.3, and 0.6–0.9 m, Respectively sampling depth (m)
EETF
ZZTFA + ZETFA
.
ZETF
ZZTF
0–0.15 (1)a 0–0.15 (2) 0–0.15 (3) 0.15–0.3 (1) 0.15–0.3 (2) 0.15–0.3 (3) 0.6–0.9 (1) 0.6–0.9 (2) 0.6–0.9 (3)
13.5 10.4 10.5 8.6 9.9 16.5 23.9 36.1 45.1
NDb ND ND ND ND ND ND ND ND
192.0 142.0 141.7 50.3 53.4 183.8 55.2 54.2 55.1
ND ND ND ND ND ND ND ND ND
ND ND ND ND ND ND ND ND ND
a
Numbers 1-3 in parentheses imply triplicate samples. b Not detected.
internal standard was taken into calculation. Because both of them were suffering from ion suppression, the moderate matrix effect was even more compensated. For this reason, the matrix effect was considered to be negligible in this study and the value of peak area from the standard solution prepared without a matrix was used to calculate the standard curve in the quantitative analyses. The fact that the observed matrix effect was insignificant could be due to the absence of any sample concentration, so that the coelutes’ interferences were not strong enough to cause serious matrix effects. No implementation of any concentration technique also simplified the whole sample preparation procedure compared to previous methods. Validation of the Method. The reliability of the method was confirmed by the spiked recovery test. The internal standard, d3-EETF, was tested first and confirmed that no EETF and its isomers were observed in the chromatography. The LOD, defined as signal-to-noise ratio (S/N) equal to 3, was 0.1 µg L-1, whereas the developer’s method reports it at 1 µg L-1 (16). The LOQ, defined as S/N ) 10, was 0.5 µg L-1. Considering the final extract volume was 50 mL and 10 g of soil was used in the analysis, the method LOQ should be 2.5 ppb in the soil. Two spike levels of recovery tests (0.05 and 3 mg kg-1) were implemented. After spiking solution was fortified directly into each 10 g soil sample in the ASE cell, the cell was placed under the hood for at least 10 min and it was subjected to the procedure described in the earlier part of this paper. For the high-level spike recovery test, another 600 times dilution step (than the normal procedure) was done to bring the concentration down to the range of calibration curve. The results are presented in Table 8. The recovery was improved from the initial 20–30% to 68.5–107.6% for the W, M, K, and P soils of Hawaii and to 53.5–72.3% for L soil if the ASE temperature stayed at 40 °C (Table 8). No more effort was put into this work to lower the spike recovery levels, like the developer’s method, because our method was used for overcoming the low recovery of Hawaii soil when applied to a leachability study, rather than the residue study. Table 9 shows linearity information of the calibration curves, which include equation parameters and correlation coefficients (r2) of each target compound. To confirm that the ASE would not lead to degradation of EETF, pure EETF (1 mL of 0.5 µg mL-1 solution) was spiked to 10 g of soil and extracted. No isomers or acid metabolites were discovered in the extraction solution. Analysis of Field Soil Samples and Degradation Soil Samples. Along with other insecticides tested, trifloxystrobin was applied at a rate of 3.12 g per plot (8 oz/acre application
1836
J. Agric. Food Chem., Vol. 56, No. 6, 2008
rate). Each plot was 6.1 m by 9.1 m. Assuming a water penetration depth of 2.5 cm during pesticide application and a bulk density of 1.1 g cm-3 (typical of structured soils in Hawaii), the concentration in the top 2.5 cm soil would be 8.79 mg kg-1. For a 5.0 cm water penetration depth, the concentration in the soil would be 4.4 mg kg-1. On the basis of these estimated concentrations, which are expected to be encountered on the day of or 1 day after the application of trifloxystrobin, the high recovery level in this study was set at 3 mg kg-1. Leaching of trifloxystrobin and other compounds of interest was evaluated over a period of 16 weeks. After application, the plots were covered with straw mulch. The total amount of rain and irrigation water was similar to the evapotranspiration demand in the study areas. Details of the study design, sampling, and results can be found in Ray et al. (20). Soil samples were retrieved from the field in weekly, biweekly, and monthly intervals in the 16 week study period. At week 16, the residual EETF concentrations ranged from 10 to 20 ng g-1 in the sampled depths (0.75-2.1 m) depending on the site. The parent compound degraded within a few days of application and produced significant amount of EETFA, which steadily degraded over the 16 week sampling period to about 1/10 of the initial values. Table 10 shows the concentration profile of EETF and its metabolites for samples collected to a depth of 2.1 m at the two test plots of the Poamoho (P) site. Laboratory aerobic degradation studies were conducted using soils from the depth intervals of 0–0.15, 0.15–0.3, and 0.6–0.9 m from the five sites over a period of 16 weeks. As the concentration of trifloxystrobin decreases with depth in the soil profile, the applied concentrations to the lower two soil layers were progressively less than that applied to the top soil layer. The moisture contents of the samples during incubation were kept similar to the field-capacity moisture contents of the soils. The soil samples were analyzed in 1-4 week intervals for EETF and its metabolites. Results showed rapid degradation of EETF (within 2 weeks) producing EETFA. Table 11 presents the concentrations of EETF and its metabolite after 16 weeks of degradation for the P soils collected from the depth intervals of 0–0.15, 0.15–0.3, and 0.6–0.9 m, respectively. As can be found from this table (as well as Table 10), EETFA is the most prominent of the four metabolites in the field leaching and laboratory degradation studies. Additional results of the fate of these compounds will be discussed in a later paper. The automated extraction (ASE) of trifloxystrobin and its metabolites from five soils of Hawaii followed by their LCMS/MS analysis using our developed method was successful and optimal. The extraction solvent ratio was optimized to maximize the recovery of the compounds, and it contained 50% methanol and 50% water with 0.5 g L-1 of EDTANa2. Addition of EDTANa2 to the extract solvent was helpful in recovery of the analytes, which were assumed to form complexes with the metal ions present in some of the Hawaii soils. This was verified for those soils in which addition of EDTANa2 enhanced extraction of the target compounds compared to the cases when no EDTANa2 were added. For some of the clayey soils, extract temperature positively correlated with extraction efficiency. The method avoids the laborious and time-consuming sample preparation steps as presented in earlier literature, which also resulted in an insignificant matrix effect of this method. The ESI soft source fragmentation pattern was found to be different from previous work for some of the compounds.
Chen et al. ACKNOWLEDGMENT
We thank Bayer CropScience (Kansas City, MO) for providing reference standards. LITERATURE CITED (1) U.S. EPA. Pesticide Fact Sheet: Trifloxystrobin, 1999. (2) Banerjee, K.; Ligon, A. P.; Schuermann, M.; Preut, H.; Spiteller, M. (E,E)-a-(Methoxyimino)-2-[({1-[3-(trifluoromethyl)phenyl]ethylidene}amino)oxymethyl]benzeneacetic acid. Acta Crystallogr., Sect. E: Struct. Rep. Online 2005, E61 (6), o1528–o1529. (3) Banerjee, K.; Ligon, A. P.; Spiteller, M. Environmental fate of trifloxystrobin in soils of different geographical origins and photolytic degradation in water. J. Agric. Food Chem. 2006, 54, 9479–9487. (4) Abreu, S. d. M.; Caboni, P.; Cabras, P.; Garau, V. L.; Alves, A. Validation and global uncertainty of a liquid chromatographic with diode array detection method for the screening of azoxystrobin, kresoxim-methyl, trifloxystrobin, famoxadone, pyraclostrobin and fenamidone in grapes and wine. Anal. Chim. Acta 2006, 573/574, 291–297. (5) Lehotay, S. J.; De Kok, A.; Hiemstra, M.; Van Bodegraven, P. Validation of a fast and easy method for the determination of residues from 229 pesticides in fruits and vegetables using gas and liquid chromatography and mass spectrometric detection. J. AOAC Int. 2005, 88 (2), 595–614. (6) Hirahara, Y.; Kimura, M.; Inoue, T.; Uchikawa, S.; Otani, S.; Haganuma, A.; Matsumoto, N.; Hirata, A.; Maruyama, S.; Iizuka, T.; Ukyo, M.; Ota, M.; Hirose, H.; Suzuki, S.; Uchida, Y. Validation of multiresidue screening methods for the determination of 186 pesticides in 11 agricultural products using gas chromatography (GC). J. Health Sci. 2005, 51 (5), 617–627. (7) De Melo Abreu, S.; Correia, M.; Herbert, P.; Santos, L.; Alves, A. Screening of grapes and wine for azoxystrobin, kresoximmethyl, and trifloxystrobin fungicides by HPLC with diode array detection. Food Addit. Contam. 2005, 22 (6), 549–556. (8) Likas, D. T.; Tsiropoulos, N. G.; Miliadis, G. E. Rapid gas chromatographic method for the determination of famoxadone, trifloxystrobin and fenhexamid residues in tomato, grape and wine samples. J. Chromatogr., A 2007, 1150 (1–2), 208–214. (9) Ueda, N.; Ibuki, S.; Uno, M.; Sowa, Y. Simultaneous determination of residual pesticides in agricultural products by SFE and GC/MS. Nara-ken Hoken Kankyo Kenkyu Senta Nenpo 2005, 39, 66–73. (10) Fujimoto, T.; Taniguchi, M.; Miyata, S. Introduction of simultaneous tests for pesticide residues accompanied by positive list system for pesticides, etc. Amagasaki-shiritsu Eisei Kenkyushoho 2006, 31, 23–27. (11) Omote, M.; Harayama, K.; Sasaki, T.; Mochizuki, N.; Yamashita, H. Analysis of simultaneous screening for 277 pesticides in malt and beer by liquid chromatography with tandem mass spectrometry. J. Am. Soc. Brew. Chem. 2006, 64 (3), 139–150. (12) Wimmer, S.; Gartner, M. Screening method for the qualitative detection of pesticides in organic crops by SPDE/GC-MS. Ernaehrung (Vienna, Austria) 2005, 29 (9), 359–367. (13) Christensen, H. B.; Granby, K. Method validation for strobilurin fungicides in cereals and fruit. Food Addit. Contam. 2001, 18 (10), 866–874. (14) Walorczyk, S.; Gnusowski, B. Fast and sensitive determination of pesticide residues in vegetables using low-pressure gas chromatography with a triple quadrupole mass spectrometer. J. Chromatogr., A 2006, 1128 (1–2), 236–243. (15) Novartis Crop Protection, I. Analytical method for the determination of CGA-279202 and its metabolites CGA-357261, CGA357262, CGA-331409, CGA-373466 and CGA-321113 in soil by high performance liquid chromatography with UV detection including validation data; Method AG-654A; Greensboro, NC, May 17, 1996.
Trifloxystrobin and Its Metabolites in Hawaii Soils (16) Novartis Crop Protection, I. Analytical method AG-683 for the determination of CGA-279202 and its metabolites CGA-373466, CGA-321113, CGA-357261, CGA-357262 and CGA-331409 in soil and thatch-sod and CGA-279202 and its metabolites CGA321113 in grass by liquid chromatography with mass spectrometric detection, including validation data; Method AG-683; Greensboro, NC, 1998. (17) Kaushik Banerjee, A. P. L. a. M. S. Photoisomerization Kinetics of trifloxystrobin. Anal. Bioanal. Chem. 2005, 382 (7), 1527– 1533. (18) Banerjee, K.; Ligon, A. P.; Spiteller, M. Spectral elucidation of the acid metabolites of the four geometric isomers of trifloxystrobin. Anal. Bioanal. Chem. 2007, 388 (8), 1831–1838. (19) Matuszewski, B. K.; Constanzer, M. L.; Chavez-Eng, C. M. Matrix effect in quantitative LC/MS/MS analyses of biological fluids: a method for determination of finasteride in human plasma at
J. Agric. Food Chem., Vol. 56, No. 6, 2008
1837
picogram per milliliter concentrations. Anal. Chem. 1998, 70, 882– 889. (20) Ray, C.; Sanda, M.; Dusek, J.; Loo, B.; Pavelkova, H.; Sobotkova, M.; Lichwa, J. Leaching of selected pesticides in Hawaii soils as influenced by soil properties and hydrologic conditions: Field and laboratory eValuations; draft final report; Water Resources Research Center, University of Hawaii; Honolulu, HI, 2007; p 69. Received for review May 24, 2007. Revised manuscript received November 21, 2007. Accepted December 18, 2007. Financial support for this project was provided by the Hawaii Department of Agriculture. J.C. worked with C.R. as a postdoctoral associate during this research. This is contributed paper CP-2007-00 of the Water Resources Research Center, University of Hawaii.
JF071527Z